Found one medium-priority bug: `getCreaturesOnCooldown()` doesn’t include pending (undelivered) revelations, and the LLM sees cooldown creatures’ names in `petSummary` causing it to ignore the VALID names list — resulting in 14 wasted LLM calls for rejected Kron revelations. Fix by including pending revelations in the cooldown set AND adding explicit “DO NOT pick” instructions to the prompt. The simulation is otherwise healthy with diverse encounters, varied conversation topics, working romance/beliefs/jobs, and good drama cycling.
(MEDIUM): `getCreaturesOnCooldown()` Doesn’t Include Pending Revelations — LLM Picks Cooldown Creatures Via petSummary Leakage
Fix: Fix Part A — `world/src/server/belief/revelation-manager.ts`, modify `getCreaturesOnCooldown()` to also include pending revelations: Fix Part B — `world/src/server/world-brain.ts`, exclude cooldown creatures from `petSummary` in the prompt so the LLM never sees their names when revelation context…
(LOW): Negative API Response Time Display Bug
Fix: Not worth addressing — purely cosmetic and doesn’t affect functionality. —
(LOW): Metrics Knowledge Log Line Spam
Fix: In the metrics reporting function, only log the Knowledge line when values change from the previous report. Alternatively, reduce the log frequency to once per 5 minutes instead of every metrics cycle. This is very minor log noise and not worth a code change on its own. — CHANGES_PROPOSED: yes…